The web survey design guidelines for Census Bureau data collections are based on the “spectrum of evidence” that exists in the absence of publicly available industry standards for all the topics covered. The spectrum ranges from:

• replicated, rigorous, randomized split ballot test results;
• conventional usability heuristics;
• usability testing results of current designs; 
• and finally, the judgement of practitioners in the field due to the lack of any other available research.

Included within each of the sections is a description of the current design, corresponding evidence to support the design, an example image when appropriate, and references.  

• The overall look and feel: fonts, colors, the banner, and footer.
• Screen elements: features that affect all screens such as navigation or modals.
• Question components: different question types such as select one, select all that apply, grids, open-text entry, etc.
• Specific screen types: common screens such as dashboards, review screens, and submission screens.

The phone and in-person survey design guidelines for Census Bureau data collections build upon the web survey design guidelines. They reflect feedback from current field representatives and survey subject matter experts.

• Format: Each section includes a description of the current design, corresponding evidence to support the design, an example image when appropriate, and references.  
• Options for the interviewer: Includes instructions, optional text, flashcards, menu options, and helpful information to keep the interview flowing smoothly.
• Navigation: Includes rule-based ability to jump to different questions or sections to make corrections or update information.

Guidelines for the paper collection mode are currently in development.